• This topic has 1 reply, 2 voices, and was last updated August 28, 2018 by Matthew C.

Migrating from Dell Servers to Nutanix Nodes

  • In the next week we are going to be Migrating from Dell Servers using ESXi VSphere 5.5 Essentials Plus to Nutanix Nodes running VSphere 6.7 Essentials Plus.

    There are numerous ways to migrate VMs from one solution to the next which includes hours of downtime to migrate storage (we don’t have V-storage V-motion).

     

    My thought is that we can replicate our VM’s to the new Servers and then only experience a few minutes of down time as opposed to hours.

    The issue and question I am posing here is; if we replicate the VMs to the new servers how can I reverse protect them to the original DR site (not back to the old servers but to my already replicated recovery Site)?

    To provide further clarity.

    My current systems are protected on site and replicated off site to servers we own in a Co-located DataCenter. My hope is to set up my new servers on-site and then replicate them to the the new servers on-site and then do a Live failover and remove them from the old servers. I want to be able to re-protect them back to the original site without having to re-replicate all that data again (takes weeks to get them all replicated. i’ve had issues before and lost all VPGs so I have done this a few times).

    Would I be able to delete the VPGs on the protected site (keeping the seeds) and then rebuild the VPGs from the new servers using the seeded data? How do I find the right disks (most of my disks are split across different data stores from the people who set up my system prior to me being hired). I think I can export configurations?

    please help.

    Hi Daniel, you’re pretty close to already having a good solution worked out. Here’s how I would do it:

    1. Setup a 2nd set of VPG’s for your machines, with the target being your new local hardware (this does require the cloud license, fyi).
    2. Move all your VM’s to your new hardware using the Zerto “move” function. (no need to reverse sync when asked)
    3. Delete all your VPG’s on your original source (they’ll all be in an error state now anyway) and keep the VMDK files at the remote site.
    4. Setup all your VPG’s from scratch, using the leftover VMDK files as seed files, so each only has to do a delta sync.

    A few tips; depending on how many VPG’s you have, you can either manually document where each VM files resides on the remote site, or you can export it to a csv and reference it that way. Can be very helpful when setting up your VPG’s again. There is a way to export and import VPG’s, but people have varying degrees of success with that.

    This is only one way to accomplish your goal. I’m sure there are others and someone else can chime in.

    Are you migrating a vcenter to your new hardware or are you starting with a brand new vcenter too?

You must be logged in to create new topics. Click here to login